Improving Multilingual Neural Machine Translation For Low-Resource Languages: French,English - Vietnamese
This work provides strong specific gains in machine translation quality for researchers and users of low-resource language pairs, particularly French-Vietnamese and English-Vietnamese.
This paper tackles the issue of rare words in multilingual neural machine translation for low-resource language pairs, specifically French-Vietnamese and English-Vietnamese. The authors propose two strategies: dynamically learning word similarity in a shared space and augmenting rare word translation ability by updating their embeddings during training. They also leverage monolingual data to increase synthetic parallel corpora. These methods resulted in significant improvements of up to +1.62 and +2.54 BLEU points over bilingual baselines for the respective language pairs.
Prior works have demonstrated that a low-resource language pair can benefit from multilingual machine translation (MT) systems, which rely on many language pairs' joint training. This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. The first strategy is about dynamical learning word similarity of tokens in the shared space among source languages while another one attempts to augment the translation ability of rare words through updating their embeddings during the training. Besides, we leverage monolingual data for multilingual MT systems to increase the amount of synthetic parallel corpora while dealing with the data sparsity problem. We have shown significant improvements of up to +1.62 and +2.54 BLEU points over the bilingual baseline systems for both language pairs and released our datasets for the research community.